A Federated Learning Framework for Healthcare IoT devices
This addresses data privacy and resource limitations for healthcare IoT applications, but it is incremental as it builds on existing federated learning methods.
The paper tackles the problem of training deep neural networks on healthcare IoT devices with privacy and resource constraints by proposing a federated learning framework that partitions the network between devices and a server, reducing communication overhead. It achieves low accuracy loss while requiring only 0.2% of the synchronization traffic compared to vanilla federated learning.
The Internet of Things (IoT) revolution has shown potential to give rise to many medical applications with access to large volumes of healthcare data collected by IoT devices. However, the increasing demand for healthcare data privacy and security makes each IoT device an isolated island of data. Further, the limited computation and communication capacity of wearable healthcare devices restrict the application of vanilla federated learning. To this end, we propose an advanced federated learning framework to train deep neural networks, where the network is partitioned and allocated to IoT devices and a centralized server. Then most of the training computation is handled by the powerful server. The sparsification of activations and gradients significantly reduces the communication overhead. Empirical study have suggested that the proposed framework guarantees a low accuracy loss, while only requiring 0.2% of the synchronization traffic in vanilla federated learning.